A study of the vibrational characteristics of two multistory buildings

Abstract

Forced vibration tests and associated analysis of two multistory buildings are described. In one case, the dynamic properties of the building measured during the tests are compared to those predicted by simple analytical models. A three-dimensional finite element model of the second building was constructed for the purpose of evaluating the accuracy of this type of analysis for predicting the observed dynamic properties of the structure. Forced vibration tests were performed on Millikan Library, a nine-story reinforced concrete shear wall building. Measurements of three-dimensional motions of approximately 50 points on each of six floors (including the basement) were taken for excitation in the N-S and E-W directions. The results revealed a complex interaction between lateral and vertical load carrying systems in both directions The results also suggest that a significant change in the foundation response of the structure occurred in the stiffer N-S direction during the San Fernando earthquake. This phenomenon was investigated through the use of two analytical models of the building which included the effects of soil-structure interaction. The Ralph M. Parsons World Headquarters building, a twelvestory steel frame structure, was also tested. The natural frequencies, three-dimensional mode shapes, and damping coefficients of nine modes of vibration were determined. Other features of this investigation included the study of nonlinearities associated with increasing levels of response and the measurement of strain in one of the columns of the structure during forced excitation. The dynamic characteristics of the building determined by these tests are compared to those predicted by a finite element model of the structure. The properties of primarily translational modes are predicted reasonably well; but adequate predictions of torsional motions were not obtained. The comparison between measured and predicted strains suggests that estimates of stress obtained from finite element analyses of buildings should be within 25 percent of those experienced by the structure for a known excitation.